Abstracts

Rationale: Idiopathic generalized epilepsy with tonic-clonic seizures (IGE-GTCS) is a disabling epilepsy syndrome, with 30% drug-resistant patients. Electrophysiological and MRI studies suggest a key role of thalamocortical networks in its pathophysiology. MRI studies have mapped structural changes and atypical connectivity, but did not address how localized anomalies may translate into macroscale pathophysiology, particularly with respect to drug-response. Connectomics taps into whole-brain organization and decomposability into interacting communities. Here, we studied the thalamocortical system in a prospective sample of IGE-GTCS and healthy controls using community-informed connectomics. Analysis was complemented with MRI morphometry, to determine functional perturbations above and beyond structural compromise. We furthermore examined associations between network anomalies and drug-response. Methods: We studied 27 IGE-GTCS prospective patients and 27 healthy controls using resting-state fMRI and structural MRI. After 1 year of follow-up, 16 patients were classified as seizure-free and 11 were drug-resistant. We subdivided the neocortex into 333 parcels and computed cortico-cortical and thalamocortical connectomes. We grouped cortical parcels into large-scale communities and assigned thalamic voxels to cortical communities using a winner-take-all scheme. For each cortical parcel and thalamic voxel, we calculated the participation coefficient, PC, a measure of between-network connectivity, and the within-module degree, WMD, to index within-network connectivity. PC and WMD measures were z-normalized relative to their community, and we compared PC and WMD measures between IGE-GTCS and controls, and between seizure-free and drug-resistant patients. FDR adjustment corrected for multiple comparisons. Connectome analysis was used to assess connectivity within and between large-scale network communities in both cortical and thalamic subregions. In addition to comparing patients to controls, we examined associations to prospective seizure control.  Results: Community-informed connectomics demonstrated that neocortical regions expressed increased functional diversity, particularly to fronto-central and temporal parcels (Figure 1). Conversely, the thalamus showed a more constrained network embedding of the thalamus in IGE-GTCS compared to controls (Figure 2). Findings remained significant after regressing out MRI-based thalamic volume and cortical thickness, suggesting independence from structural alterations. Assessing associations between baseline connectome features and prospective drug-response, we observed more marked network imbalances in drug-resistant compared to seizure-free patients. Conclusions: Our findings suggest a pathoconnectomic substrate of IGE centered on diverging changes in cortical and thalamic connectivity. More restricted thalamic connectivity could reflect the tendency to engage in recursive thalamocortical loops, which may contribute to network hyper-excitability. Conversely, increased connectional diversity of cortico-cortical networks may relay abnormal activity to an extended bilateral cortical territory. Network imbalances related to future drug-response, suggesting potential for prognostic applications.  Funding: CIHR, FRQS, NSFC